some questions

[chao zhou]

1.with the downfolding technique,  is  the finally J  the real J bettween the atoms if considering the  orbital hybirdization ?

2.i calculate the  magnon, but it has imaginary frequency, does it mean  its magnetic order is not stable

[何旭]

Hi, 

1. In my opinion, there is no "real" J. It is an effective quantity before or after the downfolding.  The Heisenberg Hamiltonian can be seen as a downfolding of the electron Hamiltonian. 

2. The current convention is to use the input cell as the primitive cell, and the FM state at Gamma point has the energy 0 (similar to the acoustic sum rule in phonons). If the  state (say at the zone boundary) has a negative energy, it is more stable than FM. That means the structure may have an antiferromagnetic ground state. If the lowest energy is at  an incommensurate q-point, the ground state is likely to be in an spin-spiral state. 

Note that this has to be verified. As the LKAG method is from perturbation theory and is only exact near the reference state. In many cases, the state predicted from the magnon structure are still correct so it can be a good starting point for identifying the ground state. But it should be always checked by calculating the energy of the ground state predicted from the magnon band. 

Best regards,

HeXu

[chao zhou]

Thanks HeXu as you see,This is the magnon spectrum I calculated. According to what you said, it is indeed an antiferromagnetic ground state, and there are many evidences to prove it.Why does it have only 3? How to analyze it.I want to ask LKAG method, is it the principle of calculating magnon? Please ask some literature references. I would also like to ask what useful information can be seen from the magneton spectrum. Thank you very much.

[何旭]

Thanks for the results. 

- The LKAG method is one of the method for the calculation of the magnons. There are other methods like the Generalized Bloch theorem. A brief review of the different methods for calculation of the exchange parameters could be found in the introduction of our TB2J paper. 

- Magnon is a key concept in spintronics.  This RMP paper of 1958 is perhaps a good starting point https://journals.aps.org/rmp/abstract/10.1103/RevModPhys.30.1

Another book I recommend is "Spin Waves: Theory and Applications " by Stancil and Prabhakar. 

Best regards,

HeXu

[chao zhou]

hi hexu ,thank you for your help,but I still want to ask how to find the lkag method of tb2j to calculate the magnon. It seems that there is no explanation in tb2j. and my magnon is discontinuous at X. what should this mean? Can you give me some opinions?

[matthieu verstraete]

Hello Chao,

If you read these original LKAG paper you will see it is briefly summarized at the beginning of our tb2j article. 

I would suspect that the discontinuity is more in the plotting, if the q points are not chosen continuously. Quick changes in the q dispersion corresponds to long range effects in the J, so it could also be a badly converged set of wannier functions

Matthieu

--
You received this message because you are subscribed to the Google Groups "TB2J" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tb2j+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tb2j/4b68fbec-caaf-4458-ab16-525ea9f94b06n%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Xu He]

Hello Chao,

For the dis-continuity, you can find the automatically generated k-point path in the page. https://wiki.fysik.dtu.dk/ase/ase/dft/bztable.html

I guess you have a tetragonal cell and the path is "GXMGZRAZ,XR,MA ", So the X point is actually Z|X, and the M point is R|M. I updated the code to add some space in between so it will be less confusing.  Could you upgrade to version 0.6.9 and try plotting again.

Best regards,

HeXu

To view this discussion on the web visit https://groups.google.com/d/msgid/tb2j/CAB--OUyPGdy8zdsiqZ0JonCBiWQHh%3D%2BOdBw5Vt7QgBSpzZ_XCw%40mail.gmail.com.

[chao zhou]

the downfolding produces the self-exchange interaction on the same site, does it have  physical meaning?

[Xu He]

Hi,

Usually it is not so meaningful. There is a sum rule that \sum_j H_ij=0, so that H_ii = - \sum_{j/=i} H_ij.  It has been removed in the output in the latest  TB2J version.

Best regards,

HeXu

To view this discussion on the web visit https://groups.google.com/d/msgid/tb2j/99b42876-f877-4c8c-9af3-cfbb40dc6ff0n%40googlegroups.com.

[chao zhou]

hi He Xu
Going back to the problem of the primitive cell, that is to say, I use the primitive cell with only one magnetic atom in it, turn on the spin polarization, and finally can calculate the exchange effect, right? Is it also right to use a more symmetrical unit cell to set the magnetic ground state?  

zhou chao

[Xu He]

Hi,

It depends on the material. In many cases the two results are close but in other cases they are not.

Note that the Heisenberg parameter are only strictly correct near the vicinity of the reference state. To model the spin interactions with only the billinear interactions can be insufficient in some materials, i.e., the J's will differ when calculated with different reference structures.  The amplitude of the difference can be used to identify if the higher order terms are important or not.  

Best regards,

HeXu

To view this discussion on the web visit https://groups.google.com/d/msgid/tb2j/91faafb0-dd60-41cc-9fc9-bf2495d9d9ccn%40googlegroups.com.

[chao zhou]

what does the reference state mean? Is it the my experimental structure or the magnetic ordering i set?In other words, the exchange parameters can be specified without using a specific magnetic ordering.If you set a different magnetic ordering, the exchange interaction will be  different, that is to say the billinear interactions can be insufficient.Is it right?

And I also want to ask Tom's ASD ,how is it developed and when will it be accessible?